Fostering financial innovation for the poor

Transcription

1 Fostering financial innovation for the poor The policy and regulatory environment by Paul B. McGuire & John D. Conroy This paper first appeared as a chapter in Private Finance for Human Development, eds Almud Weitz and J.D. Von Pischke, published on-line by the Office of Development Studies, United Nations Development Programme. Level 2, 283 Elizabeth Street (PO Box Adelaide Street) BrisbaneQld4000Australia Tel: Fax: info@fdc.org.au

2 About the authors Paul McGuire is a consulting associate of FDC, for which he has researched issues relating to microfinance and financial systems policy, as well as the assessment of microfinance institution capacities. He was principal researcher for a major study of the policy and regulatory environment for microfinance in nine Asian countries in 1997, conducted for FDC, as well as a study of the role of central banks in microfinance, prepared for the Asian Development Bank. Mr McGuire also prepared an assessment of the status of microfinance in the small island states of the Pacific for the United Nations Development Programme in 1996, and is the author of numerous articles and monographs on microfinance. He has worked as an economist for the Australian Treasury, the Organisation for Economic Co-operation and Development, the Government of the Cook Islands, and (currently) the Queensland Treasury. Dr John Conroy is special consultant to FDC, having previously served as its executive director for nine years from He is co-author of several FDC monographs on microfinance, including those dealing with NGO-commercial bank linkages and the policy and regulatory environment for microfinance in Asia, as well as a study of the role of central banks in microfinance, prepared for the Asian Development Bank. Dr Conroy has worked for extended periods in Indonesia and Papua New Guinea, and has published on development issues in a number of sectors, including education, labour and human resources, population and family planning, human settlements and shelter, and development cooperation policy as well as microfinance. 2

3 Contents About the authors 2 Abstract Introduction Financial Innovation for the poor Recent innovations Government and donor support Setting standards for MFIs Prudential regulation and supervision Standards used by second tier institutions Self-regulation and credit rating systems Savings mobilisation Principles for framing regulations Regulation for banks Minimum capital requirements Interest rates Prudential regulation and supervision Conclusions Endnotes 3

4 Abstract Since the late 1970s there has been an enormous amount of innovation in the provision of financial services to poor clients, operating at the systemic and institutional levels as well as on financial processes and products. New sets of techniques have been developed and applied by institutions such as Grameen Bank in Bangladesh and Bank Rakyat Indonesia. But, as recent developments in Bangladesh and the Philippines have shown, there is great scope for further innovation to enable specialist microfinance institutions (MFI) to reach poor households that still do not have access to financial services, and to provide more and better products and services to those that do. The policy and regulatory environment should promote continuous innovation, as well as being open to innovation itself. Financial innovation for the poor has relied heavily on support from government and donor agencies. Reaching self-sufficiency requires a considerable input of resources for capacity building and institutional development. This is especially true for innovative institutions developing new approaches to serving the poor. There is also a strong case for funding agencies to take explicit account of the importance of innovation when deciding which institutions to support. If MFIs are to flourish they must be able to operate relatively freely without unnecessary restrictions. On the other hand, most MFIs are currently not operating on a sound basis. MFIs that do not meet basic operational standards are unlikely to be able to reach large numbers of poor clients sustainably. It is therefore important to find cost-effective ways of improving standards while at the same time minimising restrictions and encouraging innovation. This requires systemic innovation. Performance and reporting standards should focus on performance rather than the methods used, so that MFIs are able to determine their own approaches to microfinance and to be innovative. However, this is not always the case. Even in Bangladesh and the Philippines, which have relatively NGO-friendly regulatory regimes, some of the guidelines used by second tier institutions relate to the method of lending rather than performance. Further work is needed to develop systems for establishing and implementing standards for MFIs that encourage innovation, while also ensuring good performance and properly managing risk. The policy and regulatory environment is particularly important if regulated banks are to offer innovative financial services to the poor. One critical factor affecting the extent to which regulated banks will become involved in microfinance is scope to establish small banks with realistic minimum capital requirements, as in Indonesia and the Philippines. Another important factor is deregulated interest rates, which enable banks to operate microfinance programs sustainably. 4

5 Introduction Policymakers in developing countries have long recognised the importance of providing financial services to the poor and near-poor. Until the 1970s and beyond, loans for microentrepreneurs were provided primarily through government and donor-funded programs. These programs generally charged subsidised interest rates, based on the view that the poor could not afford to pay market interest rates. They also tended to suffer from politicisation, low repayment rates and high arrears. They required large and ongoing subsidies, reflecting a combination of high cost structures, low interest revenues and low repayments. Perhaps most importantly, they reached only a very small proportion of poor and near-poor households. Since the late 1970s, however, there has been an emphasis on establishing innovative financial systems able to reach poor and near-poor clients on a more sustainable basis. New approaches have been developed by specialist microfinance institutions (MFIs) such as Grameen Bank in Bangladesh, and formal financial institutions such as Bank Rakyat Indonesia (BRI). It is now accepted that poor and near-poor households can be a profitable market niche for innovative financial services. This acceptance has led to initiatives such as the establishment in 1995 of the Consultative Group to Assist the Poorest (CGAP) under the auspices of the World Bank, and the Microcredit Summit in 1997 which launched a global movement to reach 100 million of the world s poorest families with financial services by Despite these achievements, the vast majority of poor and near-poor households still do not have access to financial services other than from moneylenders or other informal sources, principally the family. Women s World Banking (1995) estimates that in most developing countries, the formal financial system reaches at a maximum the top 25% of the economically active population. And despite the growing number of MFIs, CGAP (1998) suggests that they reach less than 2% of microentrepreneurs. Fewer than 25 MFIs worldwide have reached a significant size and are commercially sustainable. Hence, the challenge to extend financial services to all sections of the community remains huge, and meeting it will require attention to a wide range of issues. This essay discusses a small subset of these issues. Specifically, it focuses on the importance of financial innovation, and in particular how the policy and regulatory environment can encourage innovation. It assumes that particular regulatory environments reflect prior policy processes, and that the reform of such environments requires good policy, consciously implemented. The essay deals with particular policy regimes that appear to have engendered regulatory frameworks that facilitate innovation. It also recognises that much policy is determined without reference to the needs of the poor, and may consequently throw up unforeseen or unintended barriers to their accessing financial services. Our discussion of these subjects is based on experiences in three countries in Asia, namely Bangladesh, Indonesia and Philippines, that have been at the forefront in developing innovative approaches to financial services for the poor. The next section discusses in more detail the nature and importance of financial innovation. The importance of appropriate support from government and donor agencies is then dealt with. Later sections consider in detail the question of regulation of MFIs, and the need to establish performance and reporting standards to encourage the development of a sustainable microfinance sector while at the same time facilitating innovation, and the regulation of banks engaged in microfinance. 5

6 Financial innovation for the poor Since the late 1970s there has been an enormous amount of innovation in the provision of financial services to poor clients. As noted above, a new set of techniques has been developed and applied by institutions such as Grameen Bank, BRI, and others. These techniques have been widely documented, and are not discussed in detail here. i While they differ between institutions, Rhyne and Otero (1994) argue that there are three key principles behind them. The first is knowing the market. The poor are willing to pay for access and convenience. Lending outlets are located near the client, application procedures are simple, and loans are disbursed quickly. Second, successful MFIs use special techniques to slash administrative costs. Simple procedures are used and approvals are decentralised. Borrower groups often handle much of the loan-processing burden. Third, they use special techniques to motivate repayments. MFIs have developed a range of techniques to ensure high repayment rates, including the use of self-selected groups in which members guarantee each other s loans, intensive motivation and supervision of borrowers, incentives for borrowers, offers of progressively larger loans and compulsory savings requirements. The new approaches have demonstrated that it is indeed possible to provide financial services to large numbers of poor clients on a sustainable basis. These new approaches have been extensively studied, refined, replicated and adapted. The Grameen Bank model in particular has been replicated in numerous countries, and the basic model has proved robust in a variety of settings. ii The same is true to a lesser extent of other approaches such as the self-help group approach and the village bank approach. Nevertheless, it is clear no one model is suitable for all situations. As noted by Von Pischke, Schneider and Zander (1997: 36): Retaining and even supporting diversity is important Successful models cannot simply be replicated in social and economic environments that may differ greatly. Thus, no single model offers a panacea. Moreover, it is clear that innovation should be continuous. It is generally accepted in the management literature that best practice organisations should adopt a client focus, and develop systems that encourage continuous improvements in products and services to serve clients better. Organisations providing financial services to the poor are no exception. They should constantly monitor the economic, social, technological and other environments to improve existing products and develop new products that serve the poor better. Recent innovations Recent developments in Bangladesh, Indonesia and the Philippines show that there is indeed a great deal of innovation in financial services for the poor. Schrieder and Heidhues (1995) have proposed a useful distinction between four different types of financial innovation: Financial system innovations innovations in the financial system as a whole, such as changes in the structure of the financial sector or in the legal and regulatory framework. Financial institution innovations changes in the structure, organisation and legal form of an institution. 6

7 Process innovations the introduction of new business processes leading to increased efficiency or market expansion. Process innovations are often associated with technological progress. Product innovations the introduction of new or modified financial services, such as new credit, savings, insurance, leasing, hire purchase or other financial products. This schema is useful for our purposes, since the fourfold classification of innovations, at the levels of system, institution, process and product, establishes a hierarchy of domains for change. These domains are ordered, moving from maximum generality at the level of systemic innovation to maximum specificity at the level of the financial product. Considering financial innovation in terms of this hierarchy helps our understanding of the respective roles of policy direction (as embodied in the legal and regulatory framework) and of market forces (as seen in the behavior of actors in the financial marketplace). The policy decisions of governments, monetary authorities and regulators may effect change and stimulate innovation within all four domains. However, the top (i.e. systemic) level is where policy exerts its greatest and potentially most fruitful influence on innovation. By contrast, market forces operate with increasing influence at the successive lower levels, within rules of the game determined primarily in the systemic domain. A striking example of financial system innovation, which had marked effects in the process and product domains, is the liberalisation of interest rates in the financial system at large, which occurred in Indonesia in Among the many far-reaching consequences of this change was the stimulus for the state sector rural bank, Bank Rakyat Indonesia (BRI), to reinvent itself through a series of process and product innovations, discussed below. These innovations were introduced within the existing institutional structure but greatly increased both the bank s profitability and its ability to provide financial services to the poor and nearpoor. Again, in 1988, Indonesia removed barriers to entry for new operators in certain categories of financial institutions, including the rural banks (known as the BPRs). With their minimum capitalisation set as low as Rp50 million (then around US$25,000) these institutions grew rapidly in numbers and outreach. With freedom to set interest rates, these small, locallybased institutions display the capacity to reach the poor within the constraints imposed by the institutional form of a regulated bank. An important example of financial institution innovation is the establishment of regulated banks by specialist MFIs. In the Philippines, one of the largest MFIs, the Center for Agriculture and Rural Development (CARD), established the CARD Rural Bank in Two smaller MFIs have also established rural banks. This institutional innovation enables MFIs to offer a wider range of financial products to clients, especially on the savings side, and offers greater protection to depositors. Another financial institution innovation in the Philippines is the lending investor. Lending investors are basically moneylenders who are licensed by the central bank and subject to minimal regulation and supervision. They are permitted to accept deposits from a maximum of 19 individuals, and are permitted to make loans to whomever they wish. There has been a rapid increase in the number of lending investors in recent years, and increased competition has led to reductions in the interest rates they charge. Interest rates are now down to around 3% per month, to the point where lending investors are starting to compete with NGOs. Some excellent examples of process innovation can be seen in the case of Association for Social Advancement (ASA) in Bangladesh. ASA uses a Grameen-type approach, but has 7

8 recently introduced significant changes to its loan processes. Since 1997, it has abolished the group guarantee. Borrowers are still organised into groups of around 20, but no longer guarantee each others loans. In the case of married women, a guarantee from the husband is generally required in case the couple separate. ASA has also eased the requirement for weekly meetings. Members are required to attend weekly meetings for their first year in the program, but after that they are no longer compulsory. These changes have not resulted in any deterioration in repayment performance to date. As mentioned above, Indonesia provides many examples of process innovation leading to improved financial services for the poor. Following interest rate liberalisation in 1983, Bank Rakyat Indonesia introduced sweeping changes in the accounting systems and staff incentives for its village level offices, the unit desa. These became individual profit centers, fully accountable for their performance and with a structure of incentives that rewarded staff achievement. Substantial and continuing investment in human resource development has produced a responsible and responsive workforce, with effective information flows and overall system profitability. Important recent examples of product innovation by MFIs in Bangladesh have been the provision of new savings and insurance products. For instance, ASA has introduced a number of new savings products since 1997: It has introduced voluntary savings. Group members are still required to save Tk10 (less than US$0.19) per week, but can make voluntary savings deposits above this amount. Moreover they can withdraw both their voluntary and compulsory savings on demand, as long as they maintain a reasonable balance with the program (around 5 to 10% of the value of their loan). By June 1998 there were 873,000 group members with total savings of Tk704 million (US$11.4 million). It has also introduced voluntary savings for non-members, known as associate members. By June 1998 there were 525,000 associate members with Tk99 million (US$1.6 million) of savings. Moreover, it has introduced long-term savings since March This program takes the form of a deposit pension scheme, with depositors making a deposit each month and withdrawing the money after five years. The annual interest rate is 9% compound. By June 1998 there were 161,000 clients with total savings of Tk30 million (US$0.5 million). Some other MFIs are also offering innovative savings products. MFIs such as Federal Savings and Safe Save and are offering daily collections of savings, with field officers visiting clients on a daily basis. By December 1997, Federal Savings had 52,000 clients. BURO, Tangail is offering a deposit pension scheme. While these savings products give rise to important regulatory issues, the very strong demand for them demonstrates that they are serving poor households in a manner that was not done previously. Some other product innovations in Bangladesh are documented by Johnson and Rogaly (1997). ACTIONAID (AA) has pioneered a hire purchase scheme for poor people. Under this program, group members can hire an asset such as a rickshaw from AA, and then convert the contract into a purchase contract. Since 1988, Delta Life Insurance has been experimenting with providing basic life insurance to rural people. More recently, other organisations have also started offering insurance products to poor clients in rural areas. Other new products are being developed for small entrepreneurs who may be above the poverty line but nevertheless do not have access to the formal financial system. Since 1992 Grameen Bank has offered a leasing program for borrowers meeting certain criteria and with additional sources of income. By December 1997, 8,411 items had been leased at an 8

9 average value of Tk 28,000 (US$460) per lease. ASA introduced individual small entrepreneur loans in 1997, and by June 1998 there were some 57,000 small entrepreneurs with individual loans. BRAC is also offering larger loans. Returning to the case of Bank Rakyat Indonesia, the process innovations described above were accompanied by substantial product innovation. With the freedom to set interest rates sufficient to cover the cost of lending, and to offer deposit rates showing a positive real return to savers, BRI radically improved the financial services available to lower-income people. It introduced the well-known Kupedes loan product with a standard 1.5% flat rate per month on the smaller loans relevant to the poor. An additional 0.5% monthly is added to each payment, refunded to the borrower after 6 months of on-time repayment. Such loans, primarily for working capital, have been very successful. Similarly, BRI introduced the Simpedes savings product, which has proved spectacularly successful in mobilising the savings of lower income people in rural Indonesia, contributing to BRI s profitability. These examples show that there is a great deal of scope for financial innovation in microfinance. Innovation is essential not only to reach the large majority of poor households that still do not have access to financial services, but also to provide more and better products and services to those that do. As noted above the four classifications of innovations, namely systemic, institutional, process and product innovations, establish a hierarchy of domains for change. In a direct sense, the remaining sections of this paper are concerned primarily with financial system innovations affecting the policy and regulatory environment for microfinance. However, the concern is to ensure a policy and regulatory environment that encourages continuous innovation in financial services for the poor. In many cases, financial system innovations are necessary to foster the institution, process and product innovations that enable the financial system to serve the poor clients on an ongoing basis. This is the central theme of the remainder of this paper. Government and donor support It is generally accepted that the provision of financial services to the poor requires active support from government and donor agencies. Rural credit markets in developing countries, especially insofar as they affect the poor, are subject to market failure. For instance, Besley (1994) identifies three distinctive features of rural credit markets, namely scarce collateral, the underdevelopment of complementary institutions, and covariant risks. These factors contribute to various market failures, in particular those related to adverse selection and moral hazard. Most commentators agree that one result is lower than optimal provision of financial services in rural areas and to poor households. iii Hence government intervention is required to encourage the provision of an optimal level of financial services. Government intervention may also be justified on distributive grounds, especially where it can be shown that financial services are an effective tool for poverty reduction. This framework suggests a particular role for government and donor agencies in encouraging innovation. For instance, Besley (1994: 44) notes that: [A] learning-based argument for intervention might hold that individuals learn from the experience of others within a country. An inefficiency might develop if individuals hang back waiting for others to try things out... An obvious role for government intervention is to subsidize early innovators. Thus experiments in institutional design, such as the Grameen Bank in Bangladesh, might serve as prime candidates for subsidization. Such arguments appear only to justify subsidizing new ventures, however, and subsidies should be phased out along the way. 9

10 Similarly, Von Pischke, Schneider and Zander (1997: 36) argue that: Innovation is essential for creation of sustainable financial institutions serving the poor... Financial innovation takes time because its result is a function of confidence and trust, as well as of risk. Donors should take a long view in their provision of support. Innovation requires delivery systems, which means that emphasis should be on the development of institutions rather than simply the implementation of programs. Zeller and Sharma (1998) make similar arguments. They also note that while some innovations will succeed, donors and governments need to recognise that many others will fail. They argue that long-term support of innovation may be the most promising direction for public policy to take in the rural financial sector. It is also clear that in practice, financial innovation for the poor has in fact relied heavily on support from government and donor agencies. While it is now generally accepted that MFIs need to become self-sufficient if they are to expand their outreach, achieving selfsufficiency takes time. To reach self-sufficiency, it is generally necessary for an MFI to realise significant economies of scale, and this requires a considerable input of resources for capacity building and institutional development. This is especially true for innovative institutions developing new approaches to serving the poor. A case in point is the Grameen Bank, which has introduced innovations at three levels, in the institutional, process and product domains. The Grameen Bank has received a great deal of support from both domestic and external sources, especially in its early years of operation. Following the success of the initial experimental project by the University of Chittagong, the Grameen Bank project was established in 1979 with support from Bangladesh Bank and later from the nationalised commercial banks. Khandker, Khalily and Khan (1995) document the financial support received by the bank between 1985 and The bank received grants from external agencies of around Tk8.5 billion (US$145 million at the end-1994 exchange rate) over this period, as well as a much larger volume of funding in the form of low interest loans. Many other MFIs in Bangladesh have also relied on donor funds, especially during the establishment phase, with around 250 MFIs currently receiving foreign donations. Such support has been critical to the development of microfinance in Bangladesh. In Indonesia, BRI built on an existing institutional structure to achieve radical innovations in processes and products. The transformation from 1983 of the unit desa system of BRI also relied heavily on subsidies. The changes included a complete revamp of the structure and operations of the unit desa system, and the introduction of new loan and savings products. CGAP (1997) notes that the Indonesian government invested considerable resources to cover operating deficits of the units until they became profitable, as well as meeting the costs of staff training and technical assistance. BRI also received significant support from external agencies. The importance of government and donor agencies is mirrored in other countries. In a study of nine countries in Asia, McGuire, Conroy and Thapa (1998) found that few if any financial institutions reaching significant numbers of poor clients have been established without the support of such funding agencies. iv There is a strong case that funding agencies should take explicit account of the need for innovation when deciding which institutions to support. As noted above, an institution using innovative approaches will generally incur higher costs than an institution replicating an existing model, reflecting the higher costs of developing new systems and the greater degree of risks involved. At the same time, more innovative institutions may generate large positive externalities, such as new or improved institutional forms, processes and products that are then adopted by other institutions. Yet the Guiding Principles for Selecting and 10

11 Supporting Intermediaries, agreed by major donor agencies in October 1995, do not make any reference to innovation, and funding agencies rarely use innovation as a criterion when deciding which institutions to support. In fact, innovation may sometimes be a disqualification to obtaining support. Arguably, highly innovative institutions such as Grameen Bank were able to obtain support in the past because they were pioneers and there were no benchmarks to compare them against. However, as microfinance grows there is a tendency to establish various donor expectations or norms that MFIs must meet to obtain funding under various government and donor programs, and in some cases these norms may discourage innovative approaches. These issues are discussed in more detail below. Setting standards for MFIs At the level of the system, the regulatory environment for microfinance institutions (MFIs) is very important if they are to develop and innovate. On the one hand, if MFIs are to flourish they should be able to operate relatively freely, and charge interest rates and fees that are sufficient to cover their costs. In fact the lack of regulation goes a long way to explaining why MFIs have been so much more successful and innovative in Bangladesh than in other countries. As noted by Carpenter (1997: 15): In Bangladesh, the unique circumstances of no regulatory oversight and a large, well-funded NGO community had resulted in the ad hoc evolution of sophisticated and innovative MFIs seeking to alleviate poverty. On the other hand, it is clear that most MFIs, whether in Bangladesh or elsewhere, are not operating on a sound basis. For instance, a major survey of MFIs conducted by Sustainable Banking with the Poor (1996) found that numerous MFIs did not have accurate accounting systems. The authors asked (p. 4): If some managers of microfinance institutions are unable to answer questions about their own costs and arrears without the help of outside experts, how can they be expected to run sustainable financial institutions? Clearly, a greater emphasis on financial monitoring and reporting using standardized accounting is needed. v Clearly, MFIs that do not meet basic operational standards are not likely to be able to reach large numbers of poor clients on a sustainable basis. The task is therefore to find cost-effective ways of improving the standards of the large numbers of MFIs that are not currently operating on a sound basis, while at the same time encouraging MFIs to be innovative and avoiding restrictions which impair their efficiency or effectiveness. This requires financial system innovations. Policymakers should avoid regulations that are unrelated to performance, such as interest rate restrictions, restrictive conditions on the use of grants and loans from donor agencies, and regulations conferring discretionary powers on government agencies. In terms of developing standards, a number of options are discussed below. Prudential regulation and supervision MFIs in Bangladesh, Indonesia and the Philippines are not subject to prudential regulation and supervision. It is generally agreed that where MFIs do not accept deposits from the general public, it is unnecessary and may even be counter-productive to subject them to formal prudential regulation and supervision by a government agency. vi 11

12 There are three major reasons for this. First, it is generally held that the objectives of prudential regulation are to preserve the integrity of the financial system and to protect small depositors. It is argued that these considerations do not arise in the case of MFIs accepting savings from their members only, as such members are generally net borrowers. Nevertheless, the distinction between members and non-members is often artificial, as not all members are borrowers at any point in time. For instance in December 1997, BURO, Tangail in Bangladesh had 45,000 members but only 26,000 loans outstanding. Moreover, the trend to MFIs offering voluntary savings services means that borrowers may have savings in excess of the value of their loans. Such members stand to lose their savings if the MFI fails. These issues are discussed in more detail in the section on savings mobilisation below. Second, in most countries, there would also be significant practical difficulties in subjecting MFIs to prudential regulation and supervision. In Bangladesh there are more than 1,000 NGOs engaged in microfinance, while in the Philippines the number is around 300. Effectively regulating and supervising all of these MFIs would place a significant administrative burden on the regulatory agency. Third, prudential regulation and supervision would not necessarily result in better performance, especially in an environment of limited administrative resources and pervasive government and regulatory failure. As noted by the Deputy Governor of the Bangladesh Bank (Khaled 1998: 6): MFIs...are operating without any regulatory control, yet embezzlements or fraudulent activities are insignificant. On the other hand, the banks are so rigidly controlled by the central bank and many other agencies. Yet the banks demonstrate gross financial indiscipline. Thus regulatory control does not ensure discipline. These points notwithstanding, there is currently considerable debate within Bangladesh on regulation of MFIs. While MFIs were relatively small it was easy to avoid these issues, but MFIs in Bangladesh now provide financial services to millions of clients, and the question of regulation has inevitably arisen. The Bangladesh Bank is preparing a report, funded by the World Bank, on regulation of MFIs. At the time of writing the report was not yet available, but it is understood that it is quite NGO friendly. Standards used by second-tier institutions In many countries, one very promising avenue for ensuring that MFIs operate on a sound basis is through the lending policies of second tier microfinance institutions. Such institutions are not regulatory agencies. But, they can require MFIs that borrow from them to meet certain performance and reporting standards. Where second tier institutions are major lenders there is a strong incentive for retail MFIs to meet their standards, and it is likely that they will have a major influence throughout the microfinance sector. Two prominent second tier microfinance institutions in Asia are Palli Karma-Sahayak Foundation (PKSF) in Bangladesh and People s Credit and Finance Corporation (PCFC) in the Philippines. Palli Karma-Sahayak Foundation has a number of guidelines for the selection of its partner organisations (POs). These cover nine areas, in particular the organisation, organisers, management, human resources, working area, field activities, performance, management information system, and accounting system. In addition, PKSF has recently introduced a rating system with five categories of POs. These determine the maximum loan the PO can obtain from PKSF. The system is very complex, with 139 indicators covering the following areas: 12

13 Viability of borrowers. Institutional viability program placement, program implementation, human resource development, monitoring and evaluation, institutional culture, financial management and internal control, and status of physical assets. Financial and economic viability financial sustainability, quality of portfolio, productivity ratios, status of microcredit fund, and financial ratio analysis. PKSF requires monthly reports from all POs, and actively follows up on these reports. It visits all POs at least four times a year, and audits each PO once a year. In addition PKSF s external auditor audits one-third of POs each year. The guidelines for selection and the rating system are designed to ensure that POs meet high standards. In this respect they have been highly successful. By February 1998, PKSF had lent a total of Tk 3,068 million (US$50 million) to 151 POs. These funds were used to provide loans to 1,069,000 final borrowers, 91% of whom were women. The loan recovery rate by PKSF from POs was 98%, and repayment rates from individual borrowers to POs were also high. Not only has the outreach of MFIs increased considerably as a result of PKSF, but it has also contributed substantially to an improved level of performance among its POs in particular and MFIs in general. Nevertheless, it is clear that some of the requirements imposed by PKSF relate to the lending methods used by the PO, rather than to actual performance in providing financial services to the poor on a sustainable basis. In this way, the requirements may reduce the scope for innovative institutions to develop new ways of reaching the poor. For instance, the requirements provide that the territory served by the organisation should be well suited to microcredit. It should have a good communications network, banking facilities and easy access to markets so that borrowers can utilise their loans profitably. If interpreted strictly, this requirement could prevent POs from getting support for programs in areas which do not meet these conditions, which are likely to be the poorest areas of the country. The guidelines for selection also require that members be organised into groups and regularly attend group meetings. Some of the performance indicators are also quite prescriptive in terms of group methods. These requirements preclude lending to MFIs that provide individual loans. While most successful microfinance programs use group approaches, there are successful examples of individual lending in other countries. Moreover, in practice PKSF has generally lent only to MFIs using groups of five and weekly meetings. PKSF is looking at the possibility of funding individual loans for small business borrowers, and allowing more flexible arrangements in size of groups and frequency of meetings. Members are also required to save regularly for a minimum of six months before they obtain a loan. In practice, PKSF has allowed some flexibility in this area. For instance, ASA has a waiting period of only twelve weeks. However, ASA decided to abandon plans to reduce the waiting period to eight weeks, because this was not permitted by PKSF. Both the guidelines for selection and performance indicators require POs to avoid overlapping with the activities of other organisations in the same area. This guideline is designed to limit competition between MFIs. While it is supported by many MFIs, particularly the smaller ones, competition can play an important role in increasing the range of products available to clients, reducing interest rates, and encouraging financial innovation. In practice, the guideline has not prevented considerable competition between MFIs in Bangladesh, and it is likely that this has contributed to some of the recent innovations discussed above. 13

14 People s Credit and Finance Corporation also has eligibility criteria and performance standards for conduits or retail lenders borrowing from it. There are 16 requirements, covering organisation and management, lending performance, financials and operating systems. As with PKSF there are five levels, with the level determining the maximum size of the loan that the conduit may obtain. Conduits at higher levels are expected to have higher outreach, repayment rates, leveraging, liquidity and viability than those at lower levels. Conduits also have to progress annually to remain in the program. Unlike PKSF, the requirements all relate to performance and apply equally to conduits using different lending methodologies. Nevertheless, it should be noted that PCFC operates two programs. The first is the Helping Individuals Realise Aspirations Through Microcredit (HIRAM), which provides funding to conduits regardless of lending methodology. The second is the ADB/IFAD program, which is earmarked for Grameen Bank replicators. The ADB/IFAD Rural Microenterprise Finance Project provides two lines of credit to PCFC for on-lending to conduits, one for loanable funds (US$34.7 million) and one for institutional development (US$7.4 million). The project document stipulates that the lending programs of conduits must satisfy eleven conditions to be eligible for funds under the project. A number of these are quite specific as to the methodology to be employed. For instance, conduits must use the group methodology and provide training not only on the principles, rules and procedures of the Grameen Bank approach, but also on value formation, microenterprise management, and livelihood schemes. Group guarantees must be used. Loan releases to members of the group must be staggered, with the release of loans to other members dependent on the repayment performance of the first borrowers. And there must be weekly staff meetings with each group center. PCFC is currently seeking to have some of these conditions relaxed. The HIRAM program has been subject to resource constraints, with PCFC having a revolving fund of only P244 million (US$4 million) for the program. Hence the ADB/IFAD program is much better funded. At end June 1998 PCFC was lending to 90 conduits, of which 68 were borrowing under HIRAM and 22 under the ADB/IFAD program. PCFC is trying to move as many conduits as possible to the ADB/IFAD program, but most do not meet the eligibility criteria. In addition, the terms and conditions of the ADB/IFAD program are much more attractive than those of the HIRAM program. HIRAM loans are for one year only, whereas ADB/IFAD loans are for seven years with a three-year grace period on principal. Both have an interest rate of 12%, a 1% front-end fee and quarterly repayments. Moreover, loans for capacity building are available at an interest rate of 3% under the ADB/IFAD program, compared with 12% under HIRAM. These factors create an incentive for conduits to adopt a relatively rigid variant of the Grameen Bank approach, and discourage innovation. Self-regulation and credit rating systems Another approach to establishing performance and reporting standards for MFIs is through self-regulation. Self-regulation may take a wide variety of forms, ranging from a voluntary code of conduct to which MFIs agree to adhere, to a rigorous licensing system administered by an apex body and backed by the force of law. In between, there is a wide variety of possibilities with different degrees of monitoring and compulsion. The feasibility of various forms of self-regulation depends on a range of factors, including the extent to which there is an apex body that can represent MFIs as a whole, the quantum of resources available for monitoring and supervision, and the availability of incentives and/or sanctions to enforce compliance. Self-regulation can be an alternative to or can supplement other mechanisms for establishing performance and reporting standards. 14

15 In the Philippines, the Coalition for Microfinance Standards was established in 1996 under USAID s Developing Standards for Microfinance Project. The Coalition is a broad-based group, representing all sections of the microfinance community. As of July 1998 it had 69 members, including leading MFIs, the central bank, PCFC, two commercial banks, government planning and regulatory bodies, research and support organisations, and donor agencies. The objective of the Coalition was to develop a set of minimum indicators to serve as a basis for evaluating the performance of MFIs. It was agreed that indicators should be few in number, and should be relevant, appropriate and verifiable. In August 1998, the Coalition agreed on a set of minimum standards and performance indicators. An institution must meet the minimum standards to be considered as a serious microfinance NGO. These cover the track record, outreach, internal control, manual of operations, management information systems and write-offs and loan losses, and should be relatively straightforward for any MFI with sound operating systems and procedures. Beyond these minimum standards, there are eight performance indicators designed to rate MFIs, as follows: Number of active clients Repayment rate (total collections during previous twelve months/amount due during previous twelve months plus amount past due) Portfolio at risk (balances of loans 30 days past due/value of total loans outstanding) Operating cost ratio (operating expenses/average loans outstanding) Operational self-sufficiency (operating income/total operating expenses) Financial self-sufficiency (operating income/adjusted total operating expenses) Equity to asset ratio (total equity/total assets) Current ratio (current assets/current liabilities) MFIs are then rated on a seven-point scale against each of the eight indicators, as follows. Table 1 Performance Indicators and Ratings Indicator / Rating Number of active 500-1,000-2,000-3,000-5,000-7,000- >10,000 clients 1,000 2,000 3,000 5,000 7,000 10,000 Repayment rate <85% 85-90% 90-93% 93-95% 95-97% 97-98% >98% Portfolio at risk >15% 13-15% 11-13% 9-11% 7-9% 5-7% <5% Operating cost ratio >40% 37-40% 34-37% 31-34% 28-31% 25-28% <25% Operating selfsufficiency <70% 70-80% 80-90% >120% 100% 110% 120% Financial selfsufficiency <50% 50-60% 60-70% 70-80% 80-90% % >100% Equity to asset ratio >50% 45-50% 40-45% 35-40% 30-35% 25-30% 15-25% Current ratio < >2.0 Source: Adapted from Philippine Coalition for Microfinance Standards (1998). The indicators are then weighted, with the number of active clients having a weighting of 30% and the other indicators each having a weighting of 10%. Finally, institutions are given an overall rating on the seven-point scale. For instance, an institution with a rating of 7 (total point score more than 6.5) is considered to have excellent performance in every area evaluated. Institutions with ratings of 1 and 2 (total point score less than 2.5) are considered to be having fundamental problems in administering their microfinance 15

16 programs, with basic weaknesses in performance indicators. An institution may also receive a Not Rated (NR) classification where it lacks basic operational information or has not satisfied the minimum standards. This approach has some very attractive features. It involves only eight performance indicators, making the system easy to apply and the results easy to interpret (provided, of course, that the MFIs meet necessary accounting and reporting standards). The indicators refer to performance only, and are not affected by the methodologies employed. As such, they encourage MFIs to adopt innovative methods that improve performance. Further, the minimum standards and performance indicators, taken together, address each of the five CAMEL vii areas as follows: Capital adequacy: equity to asset ratio. Asset quality: repayment rate, portfolio at risk. Management: track record, internal control, manual of operations and management information systems from the minimum standards. Earnings: operational self-sufficiency, financial self-sufficiency. Liquidity: current ratio. Nevertheless, some caveats are in order. In particular, the ratings measure performance, rather than the extent to which the institution meets prudential standards. For instance: The ratings reward institutions that are successful in leveraging their equity, i.e., those with low equity to asset ratios. While more sustainable institutions are undoubtedly better able to raise debt finance on private capital markets, greater leveraging implies higher risk. The ratings permit MFIs to leverage up to around six times their equity base, but it is not clear how any institutions exceeding this limit would be rated. The ratings give a 30% weighting to the number of active clients. While this may be an indicator of performance, it is not relevant in terms of prudential standards. The ratings are based on an average across the eight indicators, whereas prudential considerations require a strong performance in all areas. For instance, the ratings may overstate the strength of an institution with high outreach but relatively poor portfolio quality. Another critical issue is how to implement the standards now that they have been developed. One option is to establish an independent secretariat to conduct ratings of all MFIs. This would seem to be the most effective and transparent option, so long as the secretariat is truly independent and makes the results public. Another option is to make use of the two existing networks, the Alliance of Philippine Partners in Enterprise Development (APPEND) and the Philippine Network for the Hardcore Poor (PHILNET) to conduct ratings. At present these two networks between them account for only around 15 of the 300 MFIs in the Philippines, although these MFIs represent around 80% of total outreach. Nevertheless, they could possibly rate smaller MFIs that are not members. It is also important that there be a system of incentives and sanctions that encourages MFIs to submit to the process and strive to improve their ratings. Ideally, the ratings would be accepted by PCFC, donor agencies, commercial banks and other funding agencies as a central criterion in their funding decisions. These bodies have been involved in the development of the standards, but it remains to be seen how they will use them in practice. 16

17 Savings mobilisation One particularly difficult issue for regulators is savings mobilisation by MFIs. On the one hand, savings facilities provide a valuable service to clients, and can be important in reducing poverty. As noted above, there has been a great deal of innovation in the savings products offered by MFIs, particularly in Bangladesh. On the other hand, mobilising savings places a very strong responsibility on an institution to ensure that the savings of poor clients are not at risk. Hence there is a tension between innovation in savings products and the need for prudential standards to protect savers. In many cases, MFIs require compulsory savings as a precondition to receiving a loan. CGAP (1996) argues that such savings should probably be thought of as part of the cost of the loan, rather than as true financial intermediation requiring public intervention to protect depositors. It argues that at any point in time, the vast majority of clients with compulsory savings deposits will also have loans, and will be net borrowers. However as noted above, this is not necessarily the case. Moreover, many MFIs are now offering voluntary savings facilities to their members, and even to non-members. Clearly, there is a conflict between encouraging innovative services of the one hand, and ensuring the safety of client savings on the other. Regulators have yet to come to terms with these issues. In most countries regulators have tended not to intervene so long as MFIs have raised savings from members only, but the legal situation is ambiguous. In Bangladesh the central bank has so far not prevented ASA from raising savings from associate members, but is unlikely to take such a lenient view if smaller or less financially sound MFIs did likewise. PKSF only permits MFIs with a good track record and high standards of financial accounting to use savings in their credit programs. It would be appropriate to clarify the law in all countries. One option may be to require MFIs to establish an earmarked account for voluntary savings with a regulated financial institution, with the MFI maintaining shadow accounts for each individual client. Another option may be to link performance standards and credit rating systems explicitly to the types of savings products allowed, and the uses to which savings could be put. This is clearly an area that requires further attention. Principles for framing regulations Performance and reporting standards for MFIs are a relatively new phenomenon. As the sector expands, it is appropriate to develop benchmarks to protect clients and enable funding agencies to evaluate and compare institutions. Such standards can play an important role in encouraging MFIs to improve their performance, thereby increasing the outreach and sustainability of the sector as a whole. Standards are particularly important where MFIs mobilise savings from poor clients. At the same time, it is important that standards do not discourage innovation unnecessarily. Some discouragement is probably inevitable, particularly on the savings side where prudential standards will tend to limit the types of savings products that at least some MFIs are permitted to offer. Nevertheless, it is important that standards focus on performance rather than the methods used. MFIs should be able to determine their own approaches to microfinance and to be innovative, so long as they meet objective performance standards. The review of developments in Bangladesh and the Philippines suggests that this is not always the case. Some of the guidelines used by PKSF in Bangladesh relate to method of lending rather than performance, and may discourage innovation. This is also true of the 17